Abstract
The lifetime prevalence of shoulder pain approaches 70%, which is mostly attributable to rotator cuff lesions such as inflammation, calcific tendinitis and tears. On clinical examination, shoulder ultrasound is recommended for the detection of lesions. However, there exists inter-operator variability in diagnostic accuracy because of differences in the experience and expertise of operators. In this study, a computer-aided diagnosis (CAD) system was developed to assist ultrasound operators in diagnosing rotator cuff lesions and to improve the practicality of ultrasound examination. The collected cases included 43 cases of inflammation, 30 cases of calcific tendinitis and 26 tears. For each case, the lesion area and texture features were extracted from the entire lesions and combined in a multinomial logistic regression classifier for lesion classification. The proposed CAD achieved an accuracy of 87.9%. The individual accuracy of this CAD system was 88.4% for inflammation, 83.3% for calcific tendinitis and 92.3% for tears. Cohen's k was 0.798. On the basis of its diagnostic performance, clinical use of this CAD technique has promise.
Original language | English |
---|---|
Pages (from-to) | 2315-2322 |
Number of pages | 8 |
Journal | Ultrasound in Medicine and Biology |
Volume | 42 |
Issue number | 9 |
DOIs | |
Publication status | Published - Sept 1 2016 |
Keywords
- Computer-aided diagnosis
- Rotator cuff lesions
- Shoulder ultrasound
- Texture
ASJC Scopus subject areas
- Radiological and Ultrasound Technology
- Biophysics
- Radiology Nuclear Medicine and imaging